torch.compiler API reference

Created On: Jun 02, 2023 | Last Updated On: Apr 24, 2026

For a quick overview of torch.compiler, see torch.compiler.

compile	See torch.compile() for details on the arguments for this function.
reset	Reset the in-process compiler state.
allow_in_graph	Tells the compiler frontend (Dynamo) to skip symbolic introspection of the function and instead directly write it to the graph when encountered.
substitute_in_graph	Register a polyfill handler for a function, usually a C function from the C extension, to be used in place of the original function when inlining the original function in the graph.
assume_constant_result	This function is used to mark a function fn as having a constant result.
list_backends	Return valid strings that can be passed to torch.compile(..., backend="name").
disable	This function provides a decorator to disable compilation on a function.
set_default_backend	Set the default backend for torch.compile when no backend argument is specified.
get_default_backend	Return the current default backend for torch.compile.
set_stance	Set the current stance of the compiler.
set_enable_guard_collectives	Enables use of collectives during guard evaluation to synchronize behavior across ranks.
cudagraph_mark_step_begin	Indicates that a new iteration of inference or training is about to begin.
is_compiling	Indicates whether a graph is executed/traced as part of torch.compile() or torch.export().
is_dynamo_compiling	Indicates whether a graph is traced via TorchDynamo.
is_exporting	Indicated whether we're under exporting.
keep_portable_guards_unsafe	A common function to only keep guards that can be used in both Python and non-Python environments.
skip_guard_on_inbuilt_nn_modules_unsafe	A common function to skip guards on the inbuilt nn modules like torch.nn.Linear.
skip_guard_on_all_nn_modules_unsafe	A common function to skip guards on all nn modules, both user defined as well inbuilt nn modules (like torch.nn.Linear).
keep_tensor_guards_unsafe	A common function to keep tensor guards on all tensors.
skip_guard_on_globals_unsafe	A common function to skip guards on all globals.
skip_all_guards_unsafe	A function for skipping all guards on a compiled function.
nested_compile_region	Tells ``torch.compile`` that the marked set of operations forms a nested compile region (which is often repeated in the full model) whose code can be compiled once and safely reused.
load_cache_artifacts	Hot loads cache artifacts that were previously serialized via save_cache_artifacts
load_compiled_function	Load an aot-compiled function from a file.
save_cache_artifacts	Serializes all the cache artifacts that were created during the compilation
wrap_numpy	Decorator that turns a function from np.ndarrays to np.ndarrays into a function from torch.Tensors to torch.Tensors.